English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Several recently developed methods have the potential to harness machinelearning in the pursuit of target quantities inspired by causal inference,including inverse weighting, doubly robust estimating equations andsubstitution estimators like targeted maximum likelihood estimation. There areeven more recent augmentations of these procedures that can increaserobustness, by adding a layer of cross-validation (cross-validated targetedmaximum likelihood estimation and double machine learning, as applied tosubstitution and estimating equation approaches, respectively). While thesemethods have been evaluated individually on simulated and experimental datasets, a comprehensive analysis of their performance across ``real-world''simulations have yet to be conducted.  In this work, we benchmark multiple widely used methods for estimation of theaverage treatment effect using ten different nutrition intervention studiesdata. A realistic set of simulations, based on a novel method, highly adaptivelasso, for estimating the data-generating distribution that guarantees acertain level of complexity (undersmoothing) is used to better mimic thecomplexity of the true data-generating distribution. We have applied this novelmethod for estimating the data-generating distribution by individual study andto subsequently use these fits to simulate data and estimate treatment effectsparameters as well as their standard errors and resulting confidence intervals.Based on the analytic results, a general recommendation is put forth for use ofthe cross-validated variants of both substitution and estimating equationestimators. We conclude that the additional layer of cross-validation helps inavoiding unintentional over-fitting of nuisance parameter functionals and leadsto more robust inferences.

Evaluating the Robustness of Targeted Maximum Likelihood Estimators via Realistic Simulations in Nutrition Intervention Trials